Our knowledge of the mechanism(s) by which fever-producing agents, or pyrogens, act upon the thermoregulatory regions of the brain to raise body temperature is still incomplete. Previous studies have shown that prostaglandin E2 (PGE2), a member of a family of lipids present in most organs, functions as a fever mediator in the brain. The mechanism by which peripheral pyrogens signal the brain to induce the formation of PGE2, however, is still unknown. It is also unclear which cells in the brain synthesize this PGE2. But it is established that its generation requires the mediation of an enzyme, cyclooxygenase (COX), which exists in two isoforms, one constitutive, COX-1, present in most cells, and another slowly inducible, COX-2, elicited in certain cells, e.g., phagocytic and endothelial cells, by inflammatory (e.g., pyrogenic) stimuli. It has been suggested, therefore, that perivascular phagocytic cells and/or microvascular endothelial cells (collectively termed barrier cells) in the brain may be the targets of circulating pyrogens, inducing COX-2 and consequently, producing PGE2. A difficulty with this notion, however, is that fever is initiated following the intravenous (iv) injection of a pyrogen (e.g., endotoxin) significantly quicker that the synthesis of COX-2. Neurons, however, are special among brain cells in that they express constitutive COX-2. Based on previous data by our group and other data in the literature, we have hypothesized, therefore, that peripheral endotoxin may signal the brain initially via neural inputs that stimulate the release of norepinephrine (NE) in the region where fever is regulated, viz., the preoptic anterior hypothalamus. There, NE may rapidly activate another enzyme, nitric oxide synthase (NOS), constitutively also present in neurons, thereby causing the local formation of NO which, it is known, can activate COX. NO, in turn, can activate COX-2, yielding PGE2 and the first of the characteristically biphasic febrile response to iv endotoxin. We further speculate that the second febrile rise is due to a similar sequence, initiated, however, by a different, meanwhile synthesized brain-derived pyrogen, interleukin-1beta, which activates inducible NOS in the barrier cells. This study is designed to verify these putative pathways by analyzing the activities of the various isozymes and cells presumed to be involved. The results should help to advance our understanding of the central mechanism of fever production and may have relevance to its management in infectious disease.